Computer Vision

• FasterMLP efficient vision networks combining attention mechanisms and wavelet downsampling
• TENet: Targetness Entanglement Incorporating with Multi-Scale Pooling and Mutually-Guided Fusion for RGB-E Object Tracking
• Point spread function deconvolution using a convolutional autoencoder
• EfficientSAM: Leveraged Masked Image Pretraining for Efficient Segment Anything
• Distillation-Guided Monocular Planar Recovery Network

• Towards Efficient Convolutional Neural Networks with Structured Ternary Patterns

• MULTIMODAL SEMANTIC-AWARE AUTOMATIC COLORIZATION WITH DIFFUSION PRIOR